BOD and COD Estimation of Wastewater Based on Low Cost Sensors Using Random Forest Regression Technique
Aryuanto Soetedjo, Evy Hendriarianti, Renaldi Primaswara Prasetya, Achmad Akbar Marhananda, Amandarika Widyatamara, Muhammad Edo Prastyo, Rachmad Albi Igam, Andika Yoga Pradana, Suhaena Wisma Ernia Sindy
Abstract
The Biological Oxygen Demand (BOD) and Chemical Oxygen Demand (COD) of the wastewater should be monitored regularly to maintain the quality standard. This paper presents a method to estimate the BOD and COD of the wastewater using the machine learning technique, namely the Random Forest Regression technique. The multi-sensor system consists of the pH, Turbidity, Dissolved Oxygen (DO), and Total Dissolved Solids (TDS) sensors employed to provide the dataset to the machine learning implemented on the Raspberry Pi module. The proposed multi-sensor system uses IoT technology for storing the data in the cloud during the data collection. The experimental results showed that the proposed method achieves a high accuracy of 94.7 % and R-Squared (R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ) of 0.97 for the BOD estimation and high accuracy of 94.14 % and R-Squared (R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ) of 0.97 for the COD estimation. Furthermore, the proposed system provides a fast computation time suitable for the wastewater's real-time BOD and COD monitoring.